Collin Tokheim (@ctokheim) header
Collin Tokheim (@ctokheim) profile picture

Topics: Cancer Genomics; Data Science; Targeted Protein Degradation. Prev: @DamonRunyon Fellow, @DanaFarber / Harvard; PhD @JHUBME. Opinions=own.

Boston, MA
Joined April 2018
Photo by Collin Tokheim (@ctokheim)Photo by Collin Tokheim (@ctokheim)Photo by Collin Tokheim (@ctokheim)Photo by Collin Tokheim (@ctokheim)Photo by Collin Tokheim (@ctokheim)Photo by Collin Tokheim (@ctokheim)Photo by Collin Tokheim (@ctokheim)Photo by Collin Tokheim (@ctokheim)Photo by Collin Tokheim (@ctokheim)Photo by Collin Tokheim (@ctokheim)
While AlphaGenome's non-coding variant interpretation is the major selling point, several aspects surrounding their protein-coding variant interpretation results gives the impression that they could be gaming benchmarks to inflate hype. 1/🧵
AlphaGenome (AG) & the AVI score is NOT "AI slop taking over genomics". I will give some concrete reasons why rather than evidence-free ranting (which I also like doing occasionally 😜). 1/
3
15
2
59
14,455
If I am analyzing whole-exome sequencing (WES), it would not be unusual to have 1:100 or 1:1000 ratio. When I've benchmarked AlphaMissense in the past for identifying oncogenic mutations in cancer WES, this makes the PR-AUC values close to zero, instead of impressive looking 0.9
1
2
9
1,641
So while I have no issue with AlphaGenome creating a precomputed AVI score to make the results more accessible, much like every other variant effect predictor. The utility though depends on how truly generalizable the results are.
1
9
950
Collin Tokheim retweeted
Exciting news! Our DFCI-TPD webinar channel reached 100,000 views on YouTube! Huge thanks to all speakers who shared their expertise, and all viewers! None of this would have been possible without our exceptional team @kdonovan1008 @RPNowak @BLZerf6 @hojong_yoon @MilkaKostic
2
14
6
65
12,888
Collin Tokheim retweeted
In a packed room #ACMGMtg23 with @HeidiRehm Les Biesecker and Steve Harrison presenting the next version of variant classification work at @TheACMG @ClinGenResource AMP and CAP. Version 4 will be points based. All slides available at tinyurl.com/ACMG23
2
37
2
98
9,766
Our lab website is online @MDAndersonNews ! mdanderson.org/research/depa… We're seeking highly motivated students, postdocs, and research assistants pursuing a career in cancer immunology and/or computational biology. Don't hesitate to apply!
22
3
55
8,592
ML marches toward TPD. Degradability correlates with ubiquitination potential & E2-accessible lysines in this model. It‘s fairly well appreciated now that formation of a ternary complex is necessary but not sufficient. sciencedirect.com/science/ar…
1
19
3,205
1/ My new book 'From Cell line to Command line' is finally here divingintogeneticsandgenomic… Grab it to start transforming yourself to a computational biologist! A thread on how this book came into being: 👇🧵
30
181
6
952
276,656
Collin Tokheim retweeted
So happy to see publishing of my doctoral work on targeted protein degradation. I’m very grateful to my advisor @XShirleyLiu for being the best mentor. Thank @ctokheim for all mentoring and help. Also thank @ssroyburman and @kdonovan1008 @eric_fischer lab and all coauthors!
sciencedirect.com/science/ar… This is our last piece of work on #proteindegradation at @dfcidatascience . Congratulations to @Wubing44589261 and @ctokheim , and thanks @eric_fischer Lab for the collaboration!
5
2
1
30